HCX - Unable to unstretch network extension from Cloud Manager when NE is configured in HA mode
search cancel

HCX - Unable to unstretch network extension from Cloud Manager when NE is configured in HA mode

book

Article ID: 321663

calendar_today

Updated On:

Products

VMware HCX VMware Cloud on AWS VMware Cloud on Dell EMC

Issue/Introduction

Identify a known issue with a Network Extension workflow when triggered from the HCX Cloud Manager when the NE is configured with HA.

Symptoms:
Forcing the unstretch for a network extension from the HCX Cloud Manager when the NE is configured with HA will fail.
The Network Extension UI wizard may not provide any indication for the completion of the workflow.

Cause

This is a functionality issue on the HCX Cloud Manager for the new NE HA feature released in HCX version 4.3.0.

Resolution

This issue is fixed in HCX 4.4 version onwards.

Workaround:
In general, All Network Extension workflows should be initiated from the HCX Connector or source Cloud Manager.

When planning a disaster recovery from a complete site failure, where the HCX Connector or source Cloud Manager may be lost:
  • If the on-premises Default Gateway is a DLR, or with a source Cloud Manager, do not enable HA so the unstretch workflow will work as expected on a given standalone NE appliance.
  • If the on-premises Default Gateway is an external router(using regular DVPG), unstretching the network extension is not required and manually connecting the Cloud Tier 1 interface will provide L3 connectivity to the Cloud VMs in that segment.


Additional Information

Impact/Risks:
  • All HCX versions 4.3.X are affected.
  • Network extension service will remain active.
Note: If the on premises Default Gateway is a DLR, all traffic from the Cloud VMs will continue to be forwarded back to source via the network extension. In the event that network extension is down, even if the Cloud Tier 1 is manually connected to the segment, the Cloud VMs will not be able to reach it and may become L3 isolated.
  • MON enabled VMs will continue to operate as expected.